Joint Dartmouth-Hitchcock and Dartmouth Team Wins Society of Critical Care Medicine Award
Dec 16, 2020 | by Julie Bonette
A team of researchers from Dartmouth-Hitchcock Medical Center (DHMC), Thayer School of Engineering, and Geisel School of Medicine have received a Gold Snapshot Award from the Society of Critical Care Medicine (SCCM).
The team’s award-winning abstract, “A Multivariate Machine Learning Algorithm for Occult Hemorrhage in a Porcine Model,” utilizes non-invasive technologies to take measurements from various locations in the body in order to detect bleeding in the absence of the traditional vital sign changes that precede shock. The authors used multiple machine learning methods to effectively predict the condition, known as occult hemorrhage (OH).
In preliminary testing, the researchers’ algorithm detected 98 percent of porcine subjects with OH after being observed for 21 minutes.
“No other technologies currently available to clinicians provide the same level of accuracy in detecting OH in a low-resource ambulatory setting,” said Ryan Halter, Dartmouth professor of engineering and of surgery. “The streamlined sensors and cloud-based data analytics that we’re in the process of developing are an effort to translate this technology to those in need, such as emergency medical technicians and in-the-field military medics.”
“This award is a wonderful affirmation of the likely success of our novel approach, combining multiple sensors with sophisticated machine learning, to save lives in low-resource emergency care settings,” said co-author Vikrant Vaze, the Stata Family Career Development Associate Professor of Engineering at Dartmouth.
The team’s collaborative effort in developing and validating the novel approach to detecting OH bridges sensor design, machine learning, and clinical practice.
The abstract’s first author is Samuel Klein, research associate in the department of emergency medicine at DHMC. Other members of the winning team from DHMC and Geisel include: Norman Paradis, originator of the idea to use multivariate sensing for hemorrhage detection; Joseph M. Minichiello (Geisel ’22); Justin E. Anderson (Geisel ’22); Alexander L. Lindqwister (Geisel ’22); Karen L. Moodie; Zachary J. Wanken; and Victor Borza '18 Th'18. In addition to Halter and Vaze, the abstract’s co-authors from Dartmouth Engineering include: Ethan K. Murphy; Jonathan T. Elliott; Navid Rashedi; Yifei Sun; Alexandra Hamlin; and Elisha M. Ronzio.
The award will be presented in early 2021 as part of the 50th Annual SCCM Virtual Critical Care Congress.